1031 - Sex Bias in Mutational Landscape of Chronic Lymphocytic Leukemia: Analysis of Clinical Sequencing Data

Background. Chronic lymphocytic leukemia (CLL) is more common in males and associated with worse outcomes in males, although the reasons for this remain poorly understood, despite well-established molecular markers defining prognostic impact in CLL. This study aimed to identify potential sex-based differences in genomic data and their prognostic impact on time to first treatment (TTFT) as a primary clinical endpoint in an untreated CLL cohort.
Methods. A clinical targeted next-generation sequencing panel, called The Rapid Heme Panel (RHP), was implemented by BWH, and covered all or hotspot exons of 95 genes. The RHP was performed on a consecutive cohort of 795 treatment-naïve patients with CLL. The cohort included 475 males (59.7%) and 320 females (40.3%). All cases were diagnosed based on iwCLL guidelines. Whole blood (n=690) or bone marrow (n=105) was studied. The earliest collected sample was selected for analysis if multiple samples had been studied on a given patient. The median time from diagnosis to RHP analysis was 6.6 months (26.50 – 33.27, 95% CI). Sex-based risk and association were analyzed via Cox proportional-hazards regression. TTFT was defined as the time from diagnosis to initial treatment.
Results. The median age at diagnosis was 61 (range 20 – 88) years and displayed no differences between males and females (p=0.262; Table 1). Males were more often diagnosed at Rai stage 3-4 compared to females, albeit still infrequently (M 7.5% vs. F 3.7%, p=0.015). Males had a significantly higher incidence of ß2-microglobulin (B2M) level >3.5 mg/L (p=0.005). Unmutated IGHV (UM-IGHV), del17p and complex karyotype (CK) defined as either 3+ or 5+ abnormalities had similar incidence between males and females, while del11q had a significantly higher incidence in males (M 14.5% vs. F 9.2%, p=0.032). About half of the patients harbored unfavorable ZAP70 (ZAP70+) status, with no differences by sex. The most commonly mutated (≥10%) genes in the whole cohort were ATM (24.9%), followed by NOTCH1 (18.0%), TP53 (13.6%), and SF3B1 (12.7%). Thirty additional genes had a mutation prevalence of ≥1% and <10%, with MYD88 and SF3B1 mutations more common in male patients: M 5.3% vs. F 1.9% (p=0.016) and M 14.3% vs. F 9.4% (p=0.038), respectively. FANCL and SH2B3 mutations were more common in female patients: M 0.4% vs. F 3.1% (p=0.002) and M 0.8% vs. F 2.8% (p=0.032).
We analyzed patients with a higher mutational burden (≥2 distinct mutated genes) for co-occurrence and mutual exclusivity patterns using Fisher’s exact test (p < 0.05). For example, in males, co-occurrence was found between NOTCH1 and BCOR; as well as del17p, TP53 and CREBBP. In females, co-occurrence patterns were different: NOTCH1 and NRAS; del17p, TP53 and ASXL1. In contrast, mutual exclusivity of gene mutations between ATM and FANCL (p=0.014), and MYD88 and NOTCH1 (p=0.026) was observed only in males.
The median follow-up time for all patients was 3.91 years (range 0 – 42.8). 36.6% of males received at least first-line treatment, with a median TTFT of 7.83 yrs (6.54 – 8.81, 95% CI). In contrast, only 28.1% of females were treated, with a median TTFT of 8.62 yrs (7.96 – 13.76, 95% CI), p=0.013. In univariable (UV) analyses, male sex (p=0.041), age >65 (p=0.022), UM-IGHV (p <2.0E-16), CK3 (p=7.0e-10), each of del17p, del11q, and tri12 (p < 0.001), as well as positive ZAP70 (p=0.00012) were associated with shorter TTFT. Mutations in ASXL1 (p=0.00086), ATM (p=0.00042), BCOR (p=0.0016), IKZF3 (p=0.012), NOTCH1 (p=0.000036), RUNX1 (p=0.035), SF3B1 (p=0.000013), TP53 (p=0.00004), and XPO1 (p=0.00016) were also associated with shorter TTFT. Mutated MYD88 (p=0.029) and del13q (p=7.10e-06) were associated with prolonged TTFT in UV. In multivariable (MV) analysis evaluating factors with p < 0.05 in the UV analysis, factors identified as independently worsening TTFT include mutated TP53 +/- del17p (HR 1.90, p=0.0012), UM-IGHV (HR 2.42, p<5.43E-08), CK (HR 1.65, p=0.003), del11q (HR 1.51, p=0.191), SF3B1 (HR 1.65, p=0.0023), ASXL1 (HR 2.60, p=0.00013), and IKZF3 (HR 2.47, p=0.037).
Conclusion. In this study we distinguish a shorter TTFT in males vs females, but males also have more advanced disease at diagnosis (stage, B2M) and sex is not a significant predictor of TTFT in MV analysis. Among significant factors identified in MV analysis, small sex differences were found in del11q (p=0.028) and SF3B1 (p=0.038), with higher prevalence in males.